Adversarial Search: Game Playing

1
Adversarial Search Game Playing

• Reading Chapter 6.5-6.8

2
Games and AI

• Easy to represent, abstract, precise rules
• One of the first tasks undertaken by AI (since
  1950)
• Better than humans in Othello and checkers,
  defeated human champions in chess and backgammon,
  competitive in many other games
• Major exception Go

3
Difficulty

• Games are interesting because they are too hard
  to solve
• Chess has a branching factor of 35,35100 nodes,
  approx 10154
• Need to make some decision even when the optimal
  decision is infeasible
• Drives bounded rationality research

4
(No Transcript)
5
Deep Blue

• Kasparov vs. Deep Blue, May 1997
• 6 game full-regulation chess match (sponsored by
  ACM)
• Kasparov lost the match (2.5 to 3.5)
• Historic achievement for computer chess first
  time a computer is the best chess-player on the
  planet

6

• The decisive game of the match was Game 2, which
  left a scare in my memory we saw something that
  went well beyond our wildest expectations of how
  well a computer would be able to foresee the
  long-term positional consequences of its
  decisions. The machine refused to move to a
  position that had a decisive short-term advantage
  showing a very human sense of danger. I think
  this moment could mark a revolution in computer
  science that could earn IBM and the Deep Blue
  team a Nobel Prize. Even today, weeks later, no
  other chess-playing program in the world has been
  able to evaluate correctly the consequences of
  Deep Blues position. (Kasparov, 1997)

7
Some other thoughts
8
Types of Games

• 2 player vs. multiplayer
• Chess vs. Risk
• Zero-sum vs. general-sum
• Chess vs. an auction
• Perfect information vs. incomplete information
• Chess vs. bridge
• Deterministic vs. stochastic
• Chess vs. backgammon

9
(No Transcript)
10
(No Transcript)
11
Search Formulation

• Two players Max and Min. Max moves first
• States board configurations
• Operators legal moves
• Initial State start configuration
• Terminal State final configuration
• Goal test (for max) a terminal state with high
  utility
• Utility function numeric values for final
  states. E.g., win, loss, draw with values 1, -1, 0

12
Minimax Algorithm

• Find the optimal strategy for Max
• Depth-first search of the game tree
• Note an optimal leaf node could appear at any
  depth of the tree
• Minimax principle compute the utility of being
  in a state assuming both players play optimally
  from there until the end of the game
• Propogate minimax values up the tree once
  terminal nodes are discovered
• Eventually, read of the optimal strategy for Max

13
Represents Maxs turn Represents Mins turn
14
Size of Game Search Trees

• DFS, time complexity O(bd)
• Chess
• B35(average branching factor)
• D100(depth of game tree for typical game)
• Bd3510010154nodes
• Tic-Tac-Toe
• 5 legal moves, total of 9 moves
• 591,953,125
• 9!362,880 (Computer goes first)
• 8!40,320 (Computer goes second)
• Go, branching factor starts at 361 (19X19 board)
  backgammon, branching factor around 20X20
  (because of chance nodes)

15
Which values are necessary?
16
?-? values

• Computing alpha-beta values
• ? value is a lower-bound on the actual value of a
  MAX node, maximum across seen children
• ? value is an upper-bound on actual value of a
  MIN node, minimum across seen children
• Propagation
• Update ?,? values by propagating upwards values
  of terminal nodes
• Update ?,? values down to allow pruning

17
(No Transcript)
18
(No Transcript)
19
(No Transcript)
20
(No Transcript)
21
(No Transcript)
22
?-? pruning

• Below a MIN node whose ? value is lower than or
  equal to the ? value of its ancestor
• A MAX ancestor will never choose that MIN node,
  because there is another choice with a higher
  value
• Below a MAX node whose ? value is greater than or
  equal to the ? value of its answer
• A MIN ancestor will never choose that MAX node
  because there is another choice with a lower value

23
Effectiveness of ?-? pruning (KnuthMoore 75)

• best-case if successors are ordered best-first
• ?-? must only examine O(bd/2) nodes instead of
  O(bd)
• Effective branching factor is sqrt(b) and not b
  can look twice as far ahead.
• avg-case if successors are examined in random
  order then nodes will be O(b3d/4) for moderate b
• For chess, a fairly simple ordering function
  (e.g., captures, then threats, then forward
  moves) gets close to theoretical limit
• worst case in worst-case, ?-? gives no
  improvement over exhaustive search

24
What if ?-? search is not fast enough?

• Notice that were allowing smart ordering
  heuristics, but otherwise ?-? still has to search
  all the way to terminal states for at least a
  portion of the search space.
• What else can we do??

25
Heuristics evaluation functions

• Bound the depth of search, and use an evaluation
  function to estimate value of current board
  configurations
• E.g., Othello white pieces - black pieces
• E.g., Chess Value of all white pieces Value of
  all black pices
• Typical values from infinity (lost) to infinity
  (won) or -1,1
• ? turn non-terminal nodes into terminal leaves
• And, ?-? pruning continues to apply

26
Evaluation Functions

• An ideal evaluation function would rank terminal
  states in the same way as the true utility
  function but must be fast
• Typical to define features, make the function a
  linear weighted sum of the features

27
Chess

• F1number of white pieces
• F2number of black pieces
• F3F1/F2
• F4number of white bishops
• F5estimate of threat to white king

28
Weighted Linear Function

• Eval(s)w1F1(s)w2F2(s)wnFn(s)
• Given features and weights
• Assumes independence
• Can use expert knowledge to construct an
  evaluation function
• Can also use self-play and machine learning

29
(No Transcript)
30
(No Transcript)